Re: Microsoft finally acknowledges the security drumbeats
From: Marty Fouts (usenet-user@usa.net)Date: 02/22/02
- Next message: Dave Korn: "Re: Microsoft watching us watch DVD movies (was: Microsoft finally)"
- Previous message: Dave Korn: "Re: dumb++ security"
- Maybe in reply to: John R Pierce: "Re: Microsoft finally acknowledges the security drumbeats"
- Next in thread: Philip J. Koenig: "Re: Microsoft finally acknowledges the security drumbeats"
- Reply: Philip J. Koenig: "Re: Microsoft finally acknowledges the security drumbeats"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
From: Marty Fouts <usenet-user@usa.net> Date: 22 Feb 2002 09:49:44 -0800
"Dave Korn" <no.spam@my.mailbox.invalid> writes:
> "Marty Fouts" <usenet-user@usa.net> wrote in message
> news:ueljgyrbu.fsf@usa.net...
> > Yes, but apparently I didn't slow my response down far enough you
> > could follow it. Let me try it again: Your thesis that the reason
> > there's a large PC industry is flawed. The sharing that your talking
> > about happened just fine in hardware (and continues to do so to this
> > day) under the patent system and it was hardware far more than
> > software that made the PC industry what it is today.
>
> It is neither hardware nor software that has made the industry what
> it is today. It is *people*, and specifically, the existence of
> large numbers of people with the requisite knowledge to build and
> program computers that makes the industry. Your hypothesis that
> restricting that knowledge to a tiny elite would somehow still have
> resulted in an industry of this size is ludicrous.
It would be ludicurous. It's also not my hypothesis. My hypothesis is
that the hardware industry managed enough sharing and created enough
expertise even while using the patent system. Thus, proper use of
patents to protect IP did *not* result in a 'tiny elist' and did, at
least for hardware, do precisely what patents are supposed to do:
incentivize invention to the point where the inventors _and_ the
consumers all benefit.
> There are only a tiny number of people who know enough about VLSI
> techniques to design something of the order of an x86 cpu; that's
> why there aren't all that many different cpu families around and why
> software m'frs outnumber cpu m'frs by tens of thousands to one.
Um, no, that's not why at all. Anyone who comes out of a good MSEE
program has enough knowledge to design something on the order of an
x86 CPU, and, the last time I checked, more MSEEs graduate than MSCS
students. There are, in fact, detailed papers available on all but
the most recent design techniques, and even those tend to get shown
off at the Hot Chips conference within a year of being invented.
The small number of CPUS families floating around (although I suspect
the size of that number, which is still in the teens, if not higher,
would surprise you,) is due to a complex set of reasons involving the
economics of designing and building complex systems.
> > Is that slow enough for you to follow, or do I have to dredge up
> > all the examples of GUIs from the 60s and explain the operation of
> > Moore's law to you?
>
> Don't bother, because they're both irrelevant. The examples of GUIs
> from the 60s do not speak to how widespread the PC industry would
> have become in future years if only a few people were allowed the
> knowledge of coding techniques inherent in those guis. And Moore's
> Law, not being a law of nature but a *consequence* of the huge PC
> industry, can hardly be considered evidence for it either.
OK. So I'm still not going slow enough for you. The relevance of GUIs
and Moore's law is that the software technology was around for around
20 years before it became widely available. It became widely available
almost *entirely* because the hardware necessary to run it finally
became cheap enough that it could become widely available. The
widespread availability of software has very little to do with how
'open' the source is and everything to do with how the cheap the
hardware it runs on is.
The how-to of GUI design, by the way, was widely disseminated even in
the 60s. People like Englebart and Sutherland were at research not
development facilities and wrote and published papers about how to do
GUI design. In those days software was not considered as critical a
proprietary component as it is now and it was often widely shared.
The open-source movement really started with the IBM SHARE users group
and the first Fortran compiler, and the dozens of similar users groups
surrounding other vendor's platforms.
- Next message: Dave Korn: "Re: Microsoft watching us watch DVD movies (was: Microsoft finally)"
- Previous message: Dave Korn: "Re: dumb++ security"
- Maybe in reply to: John R Pierce: "Re: Microsoft finally acknowledges the security drumbeats"
- Next in thread: Philip J. Koenig: "Re: Microsoft finally acknowledges the security drumbeats"
- Reply: Philip J. Koenig: "Re: Microsoft finally acknowledges the security drumbeats"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|
